MOST COMPACT PURLEY 2U GPU SERVER FOR DEEP LEARNING / HPC / VDI APPLICATIONS
Designed for Deep Learning, HPC (high performance computing) and VDI (desktop virtualization) applications, PNYSRA28 series supports up to 8 double-width NVIDIA GPU boards, 2 Intel Skylake-F CPUs as well as on-CPU 100Gb/s Omni-Path networking fabric. With a large capacity for both system memory (24 DIMM slots) and local storage(8 x hot-swap 2.5" SATA 6G/SAS 12G HDD/SSD external bays + 2 x 2.5" SATA 6G/SAS 12G HDD/SSD internal bays) and two expansion slots (PCIe Gen3 x16 bus Half-length low-profile), PNYSRA28 series provides great configuration flexibility to users who seek minimum rack space and hardware overhead.
Extend Your Scale with Intel® Xeon® Scalable
PNYSRA28 series is built around Intel Skylake CPU architecture - ‘Intel® Xeon® Scalable’. This new product family provides a wealth of new options in scalability - across compute, network and storage - to deliver solutions for a range of applications.
CPU Key Features
Intel® Xeon® Scalable CPU architecture brings improved features and functionality at both the host and fabric levels. These enable performance improvements - both natively on the chip and for future extensibility through compute, network and storage peripherals.
Memory Optimization |
Storage |
Networking |
Memory speed of up to 2666MHz
Up to 12 DIMMs per socket |
Built in support for VROC
Intel Optane ready |
Intel Omni-Path ready |
Optimized Thermal Design for Efficient Energy Consumption
PNYSRA28 series is a highly energy efficient system thanks to the innovative rack design and internal placement. Given its power saving architecture, including 8 high performance 15000rpm fans and 2 top-efficiency 80+ Titanium power supply units, the PNYSRA28 series guarantees an excellent airflow and thermal properties by keeping power consumption at the lowest level possible. This combination of large computing power, high energy efficiency and compact 2U footprint allows the PNYSRA28 series to deliver a great Megaflops / Teraflops per watt performance and therefore to help the most advanced supercomputing infrastructures to significantly reduce energy bills while boosting the execution of their applications.
Performance: Optimized for High CPU TDP
The thermal design of PNYSRA28 series allows for the highest-TDP rated Intel® Xeon® Scalable CPU SKUs to help achieve top performance.
Quality: Carefully Selected Components
The passive components in PNYSRA28 series are meticulously selected, which allows the processors and the memory modules to deliver their maximum performance in a super stable operating environment. PNYSRA28 series is built with ultra long lasting components to attain the highest reliability levels.
Ease of Deployment and Management
- Tool-less Rail Kit: included as a default across all new PNYSRA28 server models
- Easy Swapping: in-house designed proprietary tool-less sleds for 2.5” SSDs/HDDs
- Multiple connector options and tools are also provided where appropriate
HIGH PERFORMANCE COMPUTING (HPC)
Challenge in HPC:
Intensive and fast algorithmic computing, such as volume rendering, data-parallel vector calculation, multi-layer data linearity transformation, molecular fragments modeling.
GPU Computing as the Solution:
DEEP LEARNING
Challenge in Deep Learning:
High data training and validation iterations for linear and non-linear complex features with multiple processing layers and parameters; rapid object detection and parsing.
GPU Computing as the Solution:
VIRTUAL DESKTOP INFRASTRUCTURE (VDI)
Challenge in VDI:
Heavy graphics workload for the CPU; lengthy transfer delay of bulky graphics data; inefficient distribution of individual graphics resources; maximizing graphics performance at minimum latency.
GPU Computing as the Solution: